Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Using AI for Abnormal Behavior Detection

Using AI for Abnormal Behavior Detection

This is a presentation I used during my talk at the IT Summit Turkey 2019.


Daron Yondem

November 14, 2019


  1. Daron Yöndem http://daron.me @daronyondem

  2. • Out of the Box Anomaly Detection for Cloud Apps

    • Anomaly Detection in the world of IoT • Anomaly Detector for Everyone
  3. • Risky IP address • Login failures • Admin activity

    • Inactive accounts • Location • Impossible travel • Device and user agent • Activity rate
  4. • Entity behavioral analytics (UEBA) • Machine learning (ML)

  5. None
  6. None
  7. • Azure IoT Edge, Azure Stream Analytics offers built-in machine

    learning based anomaly detection • AnomalyDetection_SpikeAndDip and AnomalyDetection_ChangePoint
  8. An AI service that helps you foresee problems before they

  9. None
  10. None
  11. None
  12. None
  13. None
  14. Microsoft Cloud App Security Product Page https://drn.fyi/375O6ML Anomaly Detection with

    Azure Stream Analytics and ML https://drn.fyi/33ZlpPu Online Playground for Anomaly Detector https://drn.fyi/2Qhqz5C Azure Anomaly Detector Video https://drn.fyi/2CHyl0z
  15. http://daron.me | @daronyondem Download slides here; http://decks.daron.me